INDEX

Explanations

legal or technical contexts

This neuron detects section‐header labels (uppercase metadata fields) followed by a colon (e.g. “DESCRIPTION:”, “FILE:”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

of

-1.84

</em>

-1.57

Conclusión

-1.35

resiasi

-1.34

 другими

-1.30

 opgenomen

-1.27

cambi

-1.26

pagest

-1.25

our

-1.24

 Ergän

-1.24

POSITIVE LOGITS

digans

1.43

 stesse

1.38

фіка

1.34

 same

1.32

the

1.32

↵↵

1.32

 brinquedos

1.31

 bocetos

1.31

选择

1.29

 Meilleures

1.29

Activations Density 0.088%